skills.rest
sebbsssss
sebbsssssCOMMUNITY
Explore all skills by this author →
GitHub

Clude Memory MCP

Store, retrieve, and analyze long-term cognitive memories via an MCP server.

20|4|Updated Feb 8, 2026
Search Tags:#semantic search#mcp#agent memory#cognitive architecture#long-term context#episodic memory#dream cycle
One-click install
npx skills add https://github.com/sebbsssss/clude --skill clude-memory-mcp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Clude Memory MCP
Source: https://github.com/sebbsssss/clude/tree/main
Command: npx skills add https://github.com/sebbsssss/clude --skill clude-memory-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps agents overcome the limitations of short context windows by providing a persistent cognitive memory layer that can store experiences, search them later, and run dream-cycle synthesis to consolidate and refine understanding.

Core Features & Use Cases

  • MCP tools for memory lifecycle: search (recall_memories), add new facts/events (store_memory), and inspect progress (get_memory_stats).
  • Typed memory + hybrid retrieval: recall ranks memories by relevance, importance, recency, and (when enabled) vector similarity, with optional tag and type filters.
  • Dream-cycle style thinking: supports retrospective synthesis and contradiction resolution via the same underlying memory architecture.
  • Anomaly-based lateral retrieval: use find_clinamen to surface high-importance, low-relevance memories for creative or counterfactual exploration.

Quick Start

Use the agent tool recall_memories to search for memories relevant to the topic and then (optionally) store_memory to persist the new conclusion as an episodic or semantic memory.

Frequently Asked Questions about Clude Memory MCP

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I give my AI agent persistent long-term memory using MCP?▼

Persistent long-term memory is enabled by deploying an MCP server that stores, retrieves, and ranks episodic and semantic data using tools like store_memory and recall_memories.

What is the best way to search agent memories by importance and recency?▼

Searching agent memories uses the recall_memories tool to rank recall results by relevance, importance, recency, and optional vector similarity, while supporting tag and type filters.

Can I self-host agent memory storage with Supabase or run it offline?▼

Self-hosted memory storage with Supabase or local offline operation is supported alongside a hosted mode using the CORTEX_API_KEY for flexible cognitive architecture deployment.

How does an agent consolidate understanding through a dream cycle?▼

A dream cycle consolidates understanding by performing retrospective synthesis and contradiction resolution across stored long-term memories to refine the agent's self_model and introspection.

What is find_clinamen used for in cognitive memory retrieval?▼

Find_clinamen is used for anomaly-based lateral retrieval to surface high-importance, low-relevance memories, enabling creative or counterfactual exploration within the agent architecture.

Does agent memory filtering support tracking related users and wallets?▼

Agent memory filtering supports tracking specific related_user and related_wallet tags, applying importance and decay thresholds, and optionally tracking access timestamps for memory statistics.

Related Skills

Explore More Software Engineering Skills→
copilot-ldcopilot-ld
1

libstorage

Abstract local, S3, and Supabase storage behind a unified interface.

Official
Advanced
HeadyAIHeadyAI
1

heady-dream-engine

Traverse 384D vector memory to surface novel feature proposals during idle cycles.

Official
Advanced
Aktoh-CyberAktoh-Cyber

AgentDB Vector Search

Perform semantic vector search and retrieval with AgentDB for RAG workflows.

Official
Advanced
BoraPerusicBoraPerusic

agent-memory-systems

Analyze and manage agent memory for persistent cross-session context.

Community
Advanced